ServiceNow's AI Passport Sets Runtime Rules for AI Agents
By @sharedot · · 6 pages
- AI Frontier
- AI Agents
- Agent Governance
- Servicenow
ServiceNow launched AI Passport, a cryptographically signed credential enforcing AI agents' permissions at runtime, with Nvidia as launch partner.
What happened: a signed passport for every agent
On October 15, ServiceNow announced AI Passport, a cryptographically signed credential that defines an AI agent's identity, permitted actions, accessible systems and accountable owner.
Issued through the company's AI Control Tower governance platform, the passport is designed to carry those permissions across agents, tools and vendors, with checks performed as work executes rather than only at setup.
According to SiliconANGLE, senior vice president of AI products Nenshad Bardoliwalla framed the baseline plainly: 'Without that baseline, every guardrail is just advisory.'
The credential draws on business context from ServiceNow's configuration management database, governance policies and workflow approvals to set those limits.
Why it's surprising: enforcement at the point of execution
Most agent governance today stops at identification and behavior monitoring. The surprising part of AI Passport is that it enforces limits while an agent runs.
In a SiliconANGLE demonstration, ServiceNow configured a third-party customer service agent to process refunds below $25 while disabling unrelated capabilities: an $18 refund proceeded automatically, but a $40 request was rejected at runtime because it exceeded the threshold for human review.
The approach also supports graduated responses — revoking session tokens to pause, restricting access to approved tools, or deactivating by revoking credentials, with restoration possible after a recorded explanation.
Bardoliwalla called it 'more like a dimmer switch: you monitor, you pause, you restrict, and eventually, if necessary, you deactivate.'
Nvidia integration and open specs
SiliconANGLE reports that Nvidia Corp. is a design and launch partner, linking AI Passport with its Open Agent Safety Platform, including the OpenShell secure runtime and Sentry security framework, connecting enterprise policies to enforcement inside the infrastructure where agents operate.
Nvidia's director of product marketing for cybersecurity, Itay Ozery, said the central implementation challenge is 'making your enterprise policy meaningful at the point of execution.'
ServiceNow said AI Passport uses open specifications so other governance products, identity providers and enforcement systems can issue and verify credentials, and it is intended to support third-party agents as well as ServiceNow's own.
AI Passport, shadow AI detection and AI Cost Management will be available later this quarter.
Stakes for builders and the no-code side
The governance push lands alongside EmployeeWorks updates that extend the same philosophy to everyday work.
SiliconANGLE reports that employees can build Productivity Assistants and background Productivity Routines with no coding, while the ToolSmith environment builds tools and integrations from natural-language instructions, trained on more than 75,000 APIs and using governed connectors in a preproduction environment.
GuruFocus notes the launch pairs employee empowerment with credentialing and permission systems aimed at compliance and oversight, saying ToolSmith can reduce tool creation time from days to minutes.
Both outlets confirm the announcements tackle the paired challenge of giving workers agents that act across enterprise systems while keeping them inside approved boundaries.